In recent years, with an improvement of information and communication technology (ICT) infrastructures, network services have been widely used. It is important for service providers to manage systems so as to provide services to users in a stable manner at any time. When a failure such as a disconnection or a delay occurs, it may be necessary to quickly restore an interested server or network.
In general, a single network service is provided by a plurality of servers that are connected to each other through a network. When a failure occurs, it may be necessary to determine a server or a network that is the cause of the failure. For example, a web service involves a web server that receives requests from a client device, an application (AP) server that performs a process for each of the requests in a backyard, a database server that manages data, and the like. When a disconnection or a delay occurs in the client device, it may be necessary to determine a server or a network that causes the disconnection or the delay.
One of methods for determining the location of a failure in a network is a method for analyzing traffic on a wide area network (WAN) and thereby determining whether a failure exists on the WAN or a local area network (LAN). Specifically, traffic that comes from the LAN, and traffic that comes from a router located at the boundary between the LAN and the WAN, are distinguished from each other to determine the location of the failure by comparing the quality of the traffic coming from the LAN with the quality of the traffic coming from the router.
Japanese Laid-open Patent Publication No. 2007-208328 discloses a related technique.